A virtual audience system for enhancing embodied interaction based on conversational activity
HCII'11 Proceedings of the 1st international conference on Human interface and the management of information: interacting with information - Volume Part II
Implementing expressive gesture synthesis for embodied conversational agents
GW'05 Proceedings of the 6th international conference on Gesture in Human-Computer Interaction and Simulation
Hi-index | 0.00 |
In this paper, we discuss the feasibility of estimating the activation level of a conversation by using phonetic and turn-taking features. First, we recorded the voices of conversations of six three-person groups at three different activation levels. Then, we calculated the phonetic and turn-taking features, and analyzed the correlation between the features and the activity level. The analysis revealed that response latency, overlap rate, and speech rate correlate with the activation levels and they are less sensitive to individual deviation. Then, we formulated multiple regression equations, and examined the estimation accuracy using the analyzed data of the six three-person groups. The results demonstrated the feasibility to estimate activation level at approximately 18% root-mean-square error (RMSE).